当前位置: X-MOL 学术Pattern Anal. Applic. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
ChoiceNet: CNN learning through choice of multiple feature map representations
Pattern Analysis and Applications ( IF 3.7 ) Pub Date : 2021-07-11 , DOI: 10.1007/s10044-021-01004-9
Farshid Rayhan 1 , Aphrodite Galata 1 , Tim F. Cootes 2
Affiliation  

We introduce a new architecture called ChoiceNet where each layer of the network is highly connected with skip connections and channelwise concatenations. This enables the network to alleviate the problem of vanishing gradients, reduces the number of parameters without sacrificing performance and encourages feature reuse. We evaluate our proposed architecture on three independent tasks: classification, segmentation and facial landmark localisation. For this, we use benchmark datasets such as ImageNet, CIFAR-10, CIFAR-100, SVHN CamVid and 300W.



中文翻译:

ChoiceNet:通过选择多个特征图表示进行 CNN 学习

我们引入了一种名为 ChoiceNet 的新架构,其中网络的每一层都与跳过连接和通道级联高度连接。这使网络能够缓解梯度消失的问题,在不牺牲性能的情况下减少参数数量,并鼓励特征重用。我们在三个独立的任务上评估我们提出的架构:分类、分割和面部标志定位。为此,我们使用基准数据集,例如 ImageNet、CIFAR-10、CIFAR-100、SVHN CamVid 和 300W。

更新日期:2021-07-12
down
wechat
bug